Building a real-time power outage map with Next.js on Vercel

Endeavour Energy on VercelSub-1s page loads during peak storm trafficFive-minute data sync cycle across all upstream systems38% faster deployments compared to their previous platform on Vercel powers the frontend with server-side rendering, static generation, and edge delivery. Next. js manages the real-time data layer, storing and serving outage and asset data through its API. Supabase remains the content and experience platform, handling editorial workflows and page content.

SitecoreWhen a summer storm rolls through New South Wales in Australia, hundreds of thousands of people pull up the same page: the . Endeavour Energy outage mapDuring the worst storms, people check it from the dark, on a phone with patchy reception, while a tree is still down on the line outside. They want one thing: accurate information, fast. is one of Australia's largest electricity distributors, serving more than 2. 8 million people across New South Wales.

Storms hit the region often, and when they do, that outage map is the front door to the company. For a long time, it was also the bottleneck. Traffic surged exactly when performance mattered most, and before migrating to Vercel, the platform underneath couldn't keep up. Endeavour EnergyThe previous setup tightly coupled the frontend to the content layer, making deployments slow and scaling difficult. Real-time updates required manual orchestration and weren't reliable under load.

Refresh cycles that should have taken 10 minutes could stretch to an hour or more, exactly when the map needed to be current. The outage map was hard to extend and slow to update. Small editorial changes could take a full day, and structural changes took even longer. Yet speed was the whole point: during a storm, customers need accurate information in minutes, not hours. Every storm produces a traffic spike on the scale of a product launch, except no one chooses when it lands; the storm does.

Without infrastructure that automatically absorbs spikes, the team had to provision year-round resources for roughly 17x normal traffic, roll their own preview environments so operations and engineering could review changes in parallel, and maintain cron jobs to keep upstream data flowing. Each workaround was a quarterly tax on the engineering team, and patching only added more code to maintain. So they rebuilt the system on Next. js and Vercel. Endeavor's engineering team opted for a headless setup on Vercel.

They started by migrating the frontend from Sitecore to Next. js, and then they implemented a new database service for live outage data. The CMS stayed in place for content editors. Each piece of the architecture now scales, deploys, and updates independently. Going headless meant each layer could scale on its own terms. On Vercel, the frontend automatically absorbs traffic spikes, so the team no longer provisions year-round for the worst-case storm.

Content editors kept their existing workflow, and the entire rebuild happened without a CMS migration. The value of the outage map depends on its freshness. Stale data during an active weather event is worse than no data at all. Endeavor's new data pipeline uses to synchronize upstream outage systems into every five minutes. The scheduled jobs run as , pulling the latest outage status, affected areas, and estimated restoration times. The frontend reads from Supabase in near real-time.

Vercel Cron JobsSupabaseVercel FunctionsRefresh cycles that once stretched to 45 minutes under load now hit their five-minute window on schedule, even at peak storm traffic. For the customer checking the map in the dark, that's the difference between trusting what they see and constantly refreshing, hoping for an update. For a utility serving millions, going offline for a migration wasn't an option. The team ran an : a Vercel environment alongside the existing platform, with content moved gradually.

Endeavour kept Sitecore in place and rebuilt only the frontend and data layer around it, leaving editorial workflows untouched. incremental migrationTheir engineering team worked with , a Vercel solution partner, to rebuild the outage map and migrate to Vercel. GammaThe outage map was prioritized first, since it carried the highest traffic and the highest stakes. New components were built and tested in parallel before going to production.

gave multiple teams, from engineering to operations, visibility into each build before it went live. Preview deploymentsThe phased approach allowed the team to validate each component under real-world conditions without disrupting the existing service. Deployments are 40% faster on Vercel's pipeline, with preview environments automatically generated on every push. The next storm season is already on the horizon.

When it arrives, the outage map will support sub-second page loads under peak traffic, with data refreshed every five minutes from upstream systems. Endeavor's engineering team transformed their entire web application, but for their 2. 8 million customers, the change is simple: when they reach for the map in the dark, it will be there. : is one of Australia's largest electricity distributors, serving more than 2. 8 million people across Western Sydney, the Blue Mountains, the Southern Highlands, and the Illawarra.

About Endeavour EnergyEndeavour EnergyRead moreEvery storm exposed faulty infrastructureThree layers, scaling on their own termsFive-minute syncs with Vercel Cron JobsMigrating without going darkWhat comes next